Hey everybody (3rd edit)
A recent thread about phase limiting piqued my interest.
DrDub has mentioned this monofilter plugin:
http://www.nugenaudio.com/monofilter.php
Which has been very helpful- thanks!
but I'd like to know the mechanics of how phase applies to the cutting head/stylus. I've found the RIAA groove depth limitations and the vinylrecorder site has a nice animation, but I'd like a little more in-depth understanding. Can anyone point me in a good direction?
Funny how it seems the more one learns, the more questions arise. At least I'll never run out of stuff to learn!
phasors set to 'confuse'
Moderators: piaptk, tragwag, Steve E., Aussie0zborn
-
blacknwhite
- Posts: 483
- Joined: Thu Apr 24, 2008 2:57 am
- Location: US
Re: phasors set to 'confuse'
I don't know if this will help you or not, but hope so: For a regular 45-45-type stereo cutterhead:Bratwurst wrote:Hey everybody (3rd edit)
A recent thread about phase limiting piqued my interest.
DrDub has mentioned this monofilter plugin:
http://www.nugenaudio.com/monofilter.php
Which has been very helpful- thanks!
but I'd like to know the mechanics of how phase applies to the cutting head/stylus. I've found the RIAA groove depth limitations and the vinylrecorder site has a nice animation, but I'd like a little more in-depth understanding. Can anyone point me in a good direction?
Funny how it seems the more one learns, the more questions arise. At least I'll never run out of stuff to learn!
If the source tape signal is perfect mono, i.e. the left and right channels are identical, then the cutterhead stylus movement will be nearly perfectly side-to-side movement, with no vertical component. Picture: (where the waveform would indicate the voltage of the signal from the tape player being fed into the cutterhead amp, NOT the output of the cutterhead amp, which is different)
If the signal is perfectly out-of-phase on the tape, this means the left channel's voltage is a mirror image of the right. This can happen with electronically generated music or effects, and it can also happen if, for example, the left channel mike is half of one wavelength further away from the singer than the right channel mike, for the wavelength of a certain note: Picture:
If the vertical movement is too much, then the stylus lifts out of the laquer entirely at times, tossing the playback needle into the air, resulting in random skipping.
does that help? or just raise more questions?
-
cuttercollector
- Posts: 431
- Joined: Sun Jun 11, 2006 4:49 pm
- Location: San Jose, CA
Good diagrams of in phase (mono) vs. totally out of phase 180 degrees.
It is wise to remember that this problem becomes worse as the signal gets louder, lower in frequency, and closer to being exactly out of phase.
In theory if you could record identical amplitude out of phase signals they would exactly cancel one another, one speaker cone would move in exactly as the other moved out. Of course you know if you reverse the positive and negative of one speaker and play a mono source you don't exactly hear nothing, but the bass is reduced and the sound doesn't seem like it is coming from a phantom speaker half way between the 2 any more. If, however you flip the positive and negative on one channel of a stereo playback cartridge, play a mono disc, and sum the 2 channels together as with a mono switch, you will hear very close to nothing.
The reason all of this is true is that if you think about it the left groove wall is a mirror image of the right one - exactly out of phase in a mono disc. Lateral motion only = mono compatibility when played back with a stereo cartridge. So any diference between the 2 groove walls is stereo and produces vertical modulation. Horizontal (lateral) can be thought of as the sum of the 2 channels and vertical as the diference between the 2. Or you can just call one groove wall left and the other right. Stereo will always have some information that is out of phase (that's how the old Dolby pro-logic and before that SQ records worked). The trick for cutting vinyl is to not have too much loud low frequency out of phase information as that is what will cause the groove to have excess vertical modulation pinching the playback stylus right out of the groove if it gets bad enough.
There - I'm sure THAT made everything nice and clear...
It is wise to remember that this problem becomes worse as the signal gets louder, lower in frequency, and closer to being exactly out of phase.
In theory if you could record identical amplitude out of phase signals they would exactly cancel one another, one speaker cone would move in exactly as the other moved out. Of course you know if you reverse the positive and negative of one speaker and play a mono source you don't exactly hear nothing, but the bass is reduced and the sound doesn't seem like it is coming from a phantom speaker half way between the 2 any more. If, however you flip the positive and negative on one channel of a stereo playback cartridge, play a mono disc, and sum the 2 channels together as with a mono switch, you will hear very close to nothing.
The reason all of this is true is that if you think about it the left groove wall is a mirror image of the right one - exactly out of phase in a mono disc. Lateral motion only = mono compatibility when played back with a stereo cartridge. So any diference between the 2 groove walls is stereo and produces vertical modulation. Horizontal (lateral) can be thought of as the sum of the 2 channels and vertical as the diference between the 2. Or you can just call one groove wall left and the other right. Stereo will always have some information that is out of phase (that's how the old Dolby pro-logic and before that SQ records worked). The trick for cutting vinyl is to not have too much loud low frequency out of phase information as that is what will cause the groove to have excess vertical modulation pinching the playback stylus right out of the groove if it gets bad enough.
There - I'm sure THAT made everything nice and clear...
-
cuttercollector
- Posts: 431
- Joined: Sun Jun 11, 2006 4:49 pm
- Location: San Jose, CA
Just looked at the diagrams again. I see the little dotted line which indicates you were drawing waveforms as on a scope.
Your diagram would serve just as well as the groove walls of a record with a single loud low frequency recorded. The first would be a mono groove with the 2 sides symmetrical and opposite mirror image. The 2nd one would be what would happen if you flipped the phase of a mono signal to one side of the cutter so it was trying to cut a pure vertically modulated groove. A perfect illustration of an artificially generated waveform. You could get something close to this also by recording a low frequency instrument with 2 mics that were electrically out of phase with one another, one on each channel. Think it couldnt happen in the real world? Try a Shure and an EV mic panned left and right on a bass guitar cabinet. They are electrically wired backwards from each other - or used to be !
Your diagram would serve just as well as the groove walls of a record with a single loud low frequency recorded. The first would be a mono groove with the 2 sides symmetrical and opposite mirror image. The 2nd one would be what would happen if you flipped the phase of a mono signal to one side of the cutter so it was trying to cut a pure vertically modulated groove. A perfect illustration of an artificially generated waveform. You could get something close to this also by recording a low frequency instrument with 2 mics that were electrically out of phase with one another, one on each channel. Think it couldnt happen in the real world? Try a Shure and an EV mic panned left and right on a bass guitar cabinet. They are electrically wired backwards from each other - or used to be !
-
cuttercollector
- Posts: 431
- Joined: Sun Jun 11, 2006 4:49 pm
- Location: San Jose, CA
I had forgotten about the animations on the vinylrecorder site.
One interesting one he had is the 3rd one down. He drew 2 coils one providing pure horizontal and the other pure vertical modulation.
This works too. Fairchild in fact built an early stereo cutter that was fully compatible and worked exactly this way. The horizontal modulation coil is fed the sum of the L + R channels exactly as with a mono cutter. The vertical coil is fed the difference between the 2 channels L - R ie. the "stereo" information. This produces exactly the same thing as feeding the left and right coils of a 45/45 cutter with their respective channels.
The trick to remember is that the 2 coils in a 45/45 stereo cutter while cutting a mono (lateral movement only) groove are ELECTRICALLY out of phase. One is "pushing" while the other "pulls" if you will to produce a normal lateral mono groove motion. Only in phase electrical information is cut. When the phase changes between the 2 coils and stereo information (a diference between the 2 channels) is introduced, the stylus starts to modulate vertically as well.
The old Edison discs were recorded "hill and dale" - pure vertical modulation only mono. That trick I mentioned in the last post about flipping the electrical phase of the playback cartridge on one channel then combining the 2 channels is how you can play those because then the cartridge will only deliver sound from vertical modulation - horizontal lateral modulation is exactly canceled. Again, don't get too hung up on this. If you follow the color codes on hooking up a playback cartridge the normal way the left and right hot and ground terminals are marked so the coils are actually ELECTRICALLY out of phase with each other which produces an in phase signal on both channels from a normal laterally modulated standard mono record - which has the phase mirror image on the 2 walls...
Clear as mud ?
One interesting one he had is the 3rd one down. He drew 2 coils one providing pure horizontal and the other pure vertical modulation.
This works too. Fairchild in fact built an early stereo cutter that was fully compatible and worked exactly this way. The horizontal modulation coil is fed the sum of the L + R channels exactly as with a mono cutter. The vertical coil is fed the difference between the 2 channels L - R ie. the "stereo" information. This produces exactly the same thing as feeding the left and right coils of a 45/45 cutter with their respective channels.
The trick to remember is that the 2 coils in a 45/45 stereo cutter while cutting a mono (lateral movement only) groove are ELECTRICALLY out of phase. One is "pushing" while the other "pulls" if you will to produce a normal lateral mono groove motion. Only in phase electrical information is cut. When the phase changes between the 2 coils and stereo information (a diference between the 2 channels) is introduced, the stylus starts to modulate vertically as well.
The old Edison discs were recorded "hill and dale" - pure vertical modulation only mono. That trick I mentioned in the last post about flipping the electrical phase of the playback cartridge on one channel then combining the 2 channels is how you can play those because then the cartridge will only deliver sound from vertical modulation - horizontal lateral modulation is exactly canceled. Again, don't get too hung up on this. If you follow the color codes on hooking up a playback cartridge the normal way the left and right hot and ground terminals are marked so the coils are actually ELECTRICALLY out of phase with each other which produces an in phase signal on both channels from a normal laterally modulated standard mono record - which has the phase mirror image on the 2 walls...
Clear as mud ?
Ha nice cuttercollector. That will be a good hour or so of mental visualisation to get a grasp of your posts! It makes sense, but It will take time for it to sink in.
I had been confused because it seemed like out of phase = bad, but in order to have a stereo cut you need phase differences. So now I know it is mainly high amplitude low frequency signals ~180 out of phase that are a problem. This 'clicked' for me.
Phase calculations in school have always given me trouble. (electrician apprentice)
I had been confused because it seemed like out of phase = bad, but in order to have a stereo cut you need phase differences. So now I know it is mainly high amplitude low frequency signals ~180 out of phase that are a problem. This 'clicked' for me.
Phase calculations in school have always given me trouble. (electrician apprentice)
-
Dub Studio
- Posts: 263
- Joined: Tue Jun 20, 2006 9:41 am
- Location: Bristol
- Contact:
Quick interjection, I think 'phase' is the wrong word to be using. When one channel is out of phase with another, that implies there is some sort of time delay between them, causing a phase shift. If for example a sine wave is sent through the left channel half a cycle later than the right channel, they will be out of phase, and create the illusion that they have opposite polarity, which is probably where this common misnomer comes from.
Similarly, the two channels of a 45/45 cutting head are not out of phase, they are just wired up with opposing polarity. +/- and -/+ instead of +/- and +/-. If the two channels have different waveforms, that doesn't necessarily mean they are out of phase, it just means they are different, and there could be any number of reasons for this. Of course some stereo differences are caused by phase differences, but not all! For example if one channel happens to be slightly louder than the other, there will be a stereo difference, even if its a perfectly time-aligned sine wave with no phase shift whatsoever.
Similarly, the two channels of a 45/45 cutting head are not out of phase, they are just wired up with opposing polarity. +/- and -/+ instead of +/- and +/-. If the two channels have different waveforms, that doesn't necessarily mean they are out of phase, it just means they are different, and there could be any number of reasons for this. Of course some stereo differences are caused by phase differences, but not all! For example if one channel happens to be slightly louder than the other, there will be a stereo difference, even if its a perfectly time-aligned sine wave with no phase shift whatsoever.
-
Dub Studio
- Posts: 263
- Joined: Tue Jun 20, 2006 9:41 am
- Location: Bristol
- Contact:
This is only partly true, the problem is that the lower the frequency, the greater the amplitude of the wave needs to be in order to sound as loud.Bratwurst wrote: I had been confused because it seemed like out of phase = bad
So assuming you have roughly equal levels of stereo difference up and down the frequency spectrum, and all frequencies are roughly the same loudness, then the lower the frequency, the greater the chances of skipping. If you compare say a vertical cut of 600hz and a vertical cut of 300hz, the difference is striking. 600hz cuts nicely, and 300hz looks more scary. These images (x40) were both cut at 0dB NAB.
600hz
http://www.dubstudio.co.uk/images/P7100056.JPG
300hz
http://www.dubstudio.co.uk/images/P7100054.JPG
